Details of the data sources used in this study.

Background
Circulating proteins play a critical role in rheumatoid arthritis (RA), yet few have been targeted therapeutically. This study aimed to identify novel protein targets for RA therapy.
Methods
We conducted a comprehensive proteome-wide Mendelian Randomization (MR), colocalization analysis, and summary-data-based MR (SMR) to explore potential causal relationships between plasma proteins and RA, with an overall sample size of 1,148,608. The GWAS data on plasma proteins were obtained from the FinnGen study, the UK Biobank Pharma Proteomics Project and Iceland GWAS data. Then, validation of key molecules’ differential expression pattern was done using external transcriptomic data from RA patients, while the Drug Signatures Database (DsigDB) was used to identify potential therapeutic drugs. Drugs and target proteins interactions was evaluated with molecular docking and molecular dynamics simulations approaches. Potential side effects of plasma proteins associated with RA were elucidated by phenome-wide association study (Phe-WAS) approach.
Results
Genetically predicted levels of 68 plasma proteins were associated with RA. After colocalization and SMR analysis, 6 plasma proteins (FCRL3, SUGP1, TNFAIP3, EHBP1, HAPLN4, and CILP2) have been passed all tests and identified as having potential as therapeutic targets for RA. Further Receiver operating characteristic curve (ROC) analysis indicated that three protiens (CILP2, TNFAIP3 and EHBP) have a good potential as biomarkers for RA. Differential gene analysis showed the downregulation of HAPLN4, FCRL3, EHBP1 and TNFAIP3 in RA, as well as the upregulation of CILP2 in RA. Further Phe-WAS suggested that targeting these proteins may have potential side effects.
Conclusion
Our study investigated the causal relationships between plasma proteins and RA, deepening our understanding of the molecular mechanisms and facilitating the development of new therapeutic drugs.